I won't listen to Bieber in my spaceship

You’ve probably heard the rumors of something called “iRadio1”, a streaming audio service that Apple is supposedly brewing and hoping to introduce at WWDC:

Apple strikes iRadio deal with Universal Music - CNET:

The company has reached a licensing deal with the world’s largest music label, Universal Music Group, according to people familiar with the situation.

“According to people familiar with the situation” is the most weaselly way a journalist can tacitly admit they actually have no hard sources. It’s not even as good as a “sources close to the singer”.

This particular rumor seems fairly plausible to me, though, mostly because it’s not particularly exciting or sexy — at least not in the way that the supposed “iWatch” is.

At the moment it seems like it’s big bad Sony who are holding things up. Like a lot of Apple stories to do with music, this one is mostly focused on licensing, rights to distribute and renumeration for artists and labels. The same goes for most tech stories about music. Spotify, Pandora, Google Play, Grooveshark — what are they giving back, how much do they cost, what’s their business model, what cross-section of artists do they own?

All important questions, but I think there is another interesting facet of the developing music scene which isn’t given as much importance as it should.

Music by numbers

First, let’s look at a couple of relevant statistics:

  • Gracenote, who maintain one of the largest internet databases of music data, have about 130 million tracks in their database. To put that in perspective, giving a lowball estimate of 3 minutes per song, that’s about 740 years of music.

  • There are Around 75 thousand albums released per year in the United States, Canada, UK and Japan alone. Assuming 45 minutes per album, that means that we add at least 6.4 years of music per year. The rate of creation only increases, as the population gets larger and producing music becomes easier with technology.

Whilst digital sales of music (due to increased availability) are growing at a moderate rate, adoption of streaming services is expanding dramatically.

(Paid downloads 5%, Spotify 86%, Pandora 68%) [stats via Digital Music News]

I think it’s fair to say that over the past few decades there has been a marked trend towards music listening tastes which are significantly broader (covering an increased number of songs) but less in-depth (no one listens to albums any more!)

If this is how you listen to music, then it makes a lot more sense to pay a subscription to have access to an unlimited number of songs, rather than buy a much smaller number of individual songs which you can listen to multiple times. The exploding availability of high-bandwidth connections means that streaming music on the fly from someone else’s server is so much easier than it was even five years ago.

Personally, I made the switch away from downloaded music almost completely in 2011, when I started paying for a Spotify subscription. Since then, I’ve only purchased one album.

The Future Music Business Tripod

There are three things that a purveyor of music to the masses is going to have to get right in order to maintain a successful business.

  1. What music they have to sell.

  2. How much they charge for it and how.

  3. How they present it.

It’s obvious that 1 and 2 are treated with great importance. Those who can are licensing music, and I think that that streaming is going to become ubiquitous soon.

Where everyone seems to be lagging is number 3.

The Third Leg: Presenting the Music

There are obvious concerns with the actual design of the app or interface with which you access the music. iTunes has its UI problems, so does Spotify. People get turned off by awkward interfaces. But lets assume that most big music providers in the future will be able to hire halfway decent UI designers.

The real challenge, and the one that I haven’t seen a good solution for yet, is dealing with all that music. With a vast accumulation of music rights comes a gargantuan task of, for a better word, curation.

The music providers of today, to use a dated analogy, are like a friend who has a basement full of CDs, wall to wall. You go over to his house (you don’t have a CD player, but he does) and ask to hear some good music. He insists on playing his own personal 10 song mix of popular artists, the same CD he played on repeat, the previous five times you came over. You wonder why you go over there. His mum only gives you wholegrain crackers, and his room smells kinda bad. You’re sick of hanging out with him, but you don’t have any other friends.

The music provider of the future should be like a friend who asks you what you like, then turns up half an hour later in his convertible and drives you to a cool party whilst he plays a succession of songs that you’ve never heard — but which you love. You really like a couple of the songs, so he hands over the original albums, and a six-pack of beer. Later you high-five in a totally non-ironic and awesome way.


There’s been a desultory nod towards music curation by the big players, in the form of “radio” stations based on songs that you like. These are pretty poor at best, as the algorithms they’re based on aren’t particularly clever.

The more mediation you can pass onto the lay-person the better — in the form of shared playlists, for example, a particularly low-effort form of DJing that fits today’s music-on-demand culture. Just having a giant warehouse full of music isn’t going to be enough.

In the future, the real trick will be figuring out a way to get people what they didn’t even know they wanted.

  1. What a terrible name.